The factors influencing the selection of food by flounders, Platichthys Jlesus (L.), have been investigated by analysing collections made in the Severn Estuary for 1 year and the results of experiments. Flounders measuring between 6.0 and 35 cm fed heavily on the polychaete Nereis diversicolor in February and on the amphipod Gammarus salinus between February and April. Thereafter these species were replaced by the mysid Neomysis integer with the decapod Crangon iwlgaris. Hounders shorter than 6.0 cm, fed mainly on Neomysis integer regardless of month. Numerous factors were involved in the choice of the food including (1) the maximum and minimum length of prey, (2) its spatial distribution in the water column, (3) its degree of concealment, (4) its motility and ability to escape predation, ( 5 ) conditioning of the flounders for certain foods, (6) the fish's swimming speed and (7) the turbidity and temperature of the water.Although the average length of ingested prey remained unchanged as the fish grew in size, an increase in the maximum and minimum lengths was observed. The percentage of stomachs that contained no identifiable remains in flounders greater than 6.0 cm averaged 80-95 during part of the winter and 60 % in the summer and between January and March.These changes partly reflected the influence of low water temperature on the metabolic rate and availability of prey. The dry weight of the stomach contents in flounders longer than 6.0 cm was lowest in winter but high values were recorded in the spring. This latter feature was probably because the rate of feeding was greater than that of digestion. During the summer, under more normal feeding conditions, the weight of the stomach contents remained relatively low compared to other fish populations. Flounders shorter than 6.0 cm always contained more food in their stomachs on a unit weight basis than larger individuals, reflecting metabolic rate and hunting efficiency.

Arcticchar, Sahelinusalpinus (L.), captured during the summer, 1972, in Cumberland Sound, Baffin Island, fed mainly on planktonic amphipods (Pavathenii~fo libellula, Pseudalibr-otirs glacialis), copepods (Calarius hyperboreits), and fish (Boreogadus suida, Myoxocephakus sp.), but failed to utilize coelenterates, planktonic gastropods, and epi-and infauna. The considerable variation in the species composition of the diet of char of different lengths was due primarily to size selection. The average length of all individuals in stomach contents and of representatives of most food species increased with the length of fish. The minimum and maximum length of frequently ingested organisms increased four and 90 times, respectively, as char increased from 4 to 85 cm. Char less than 10 cm in length captured in rivers tributary to the Sound fed mainly on larval Chiroiiomidae (Eickieffeviella bavaria) during the summer, whereas those longer than 10 cm fed predominantly on other char. During the winter the stomachs of the smaller individuals were always empty while the diet of the large char was restricted to other fish. Most species available to the char were consumed in proportion to their relative abundance in the rivers. The dry weight of stomach contents, when expressed on a unit weight basis, decreased with the wet weight of char in both fresh and saIt water. Thus, fish 10 g in weight contained approximately 1.5 times more food in their stomachs than those weighing 1000 g. The stomach contents of char captured in salt water weighed approximately I1 times more than those of char of comparable size captured simultaneously in fresh water. The fish fed at random intervals during the day and ceased feeding at night. Arctic char, at all sampling areas, had reached a length of approximately 9.6 crn after four years. Upon migrating to salt water, their growth rate increased sharply with the result that after eight years they were 26.5 cm in length, reflecting food availability. The growth rate gradually decreased in fish older than nine years so that 20 year olds were approximately 70.0 cm in length.

No abstract

The basis of food selection in some estuarine fishes. Eels, A nguikla anguilla (L. ) , whiting , Merlangius merlangus (L.), sprat, Sprattus sprattus (L.) and stickleback, Gmterssteus aculeatus L.The factors influencing the selection of food by eels, AnguiNa anguilla, whiting, Merlungius merlangus, sprat, Sprattus spraitus, and stickleback, Gasterosteus aculeatris have Seen investigated by analysing col!ections made in the Severn Estuary for 1 year. Non-migratory (yellow) eels measuring from 19.5-56.5 cm in length fed mainly on either the decapod Crangon vulgaris or on the mysid Ni?ornysis integer during the warmer months but ceased to feed in the winter. Whiting measuring between 2.5 and 15 cm fed almost exclusively on C. vulgaris, A' . integer and the sand goby, Pornatuschistus nzinutus. Sprats fed chiefly on eithcr the amphipod Gammarus salinus or N. integer. Sticklebacks seldom contained any food in their stomachs despite the presence of large quantities of suitable prey and the high level of hunting cfficicncy. Gammarus salinus, although normally abundant in the environment, was usually ingested in disproportionately small quantities by all fishes except sprats reflecting its concealment among floating weeds and a selection against small (less than 1 crn) organisms. Because of the limitations of size and thc unusually efficient escape reaction of larger animals, only young representatives of Crangun vulgaris were captured. The isopod Eurydice pulchra was rarely eaten by any of the fishes even though it was common in the environment. Burrowing polychaetes, mainly Nereis rldversicolor, were never utilized possibly due to the difficulty involved in removing these organisms from the substrate. Neongvsis integer was the most frequently consumed organism. Although the average length of prey ingested by the whitings increased with the size of fish, this was not the case for the other species. The proportion of non-feeding eels and the dry weight of stomach contents of this species depended on the temperature of the water but were not related to the availability of prey and tidal conditions whereas the fceding of whiting could not be related to any environmental paramter.